University of Groningen

Glucansucrases of lactobacilliKralj, Slavko

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2004

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Kralj, S. (2004). Glucansucrases of lactobacilli: characterization of genes, enzymes, and productssynthesized. s.n.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 03-06-2021

https://research.rug.nl/en/publications/glucansucrases-of-lactobacilli(e92eb104-2590-4880-9a03-d12b48259fd4).html

Chapter 3

Biochemical and molecular characterization of Lactobacillus

reuteri 121 reuteransucrase

S. Kralj, G.H. van Geel-Schutten, M.J.E.C. van der Maarel and L. Dijkhuizen

Microbiology (2004) 150: 2099-2112

Chapter 3

52

SUMMARY Lactobacillus reuteri strain 121 uses sucrose for synthesis of a unique, soluble glucan (reuteran) with mainly α-(1→4) glucosidic linkages. Previously, the gene (gtfA) encoding this glucansucrase enzyme has been characterized (Kralj et al., 2002). Here a detailed biochemical and molecular analysis of this GTFA enzyme is presented. This is believed to be the first report describing the reuteransucrase enzyme kinetics and the oligosaccharides synthesized with various acceptors. Alignments of the GTFA sequence with glucansucrases from Streptococcus and Leuconostoc origin served to identify conserved amino acid residues in the catalytic core critical for enzyme activity. Mutants Asp1024Asn, Glu1061Gln and Asp1133Asn displayed 300-1,000 fold reduced specific activities. To investigate the role of its relatively large N-terminal variable domain (702 amino acids) and its relatively short C-terminal (putative) glucan binding domain (267 amino acids, with 11 YG-repeats), various truncated derivatives of GTFA (1781 amino acids) were constructed and characterized. Deletion of the complete N-terminal variable domain of GTFA (GTFA-∆N) had little effect on reuteran characteristics (size, distribution of glucosidic linkages), but the initial transferase activity of the mutant enzyme increased drastically. Sequential C-terminal deletions (up to 6-YG repeats) in GTFA-∆N also had little effect on reuteran characteristics. However, enzyme kinetics drastically changed. Deletion of 7, 8 or 11 YG-repeats resulted in dramatic loss of total enzyme activity (43×, 63× and 1,000× fold reduced specific activities, respectively). Characterization of sequential C-terminal deletion mutants of GTFA-∆N revealed that the C-terminal domain of reuteransucrase has an important role in glucan binding. INTRODUCTION Reuteransucrase (GTFA) from Lactobacillus reuteri 121 is a 1,781 amino acid glucosyltransferase (GTF) enzyme (EC 2.4.1.5; common name glucansucrase), that synthesizes a unique soluble glucan polymer (reuteran) with mainly α-(1→4) glucosidic linkages and significant amounts of α-(1→6) and α-(1→4,6) glucosidic linkages. (van Geel-Schutten et al., 1999, Kralj et al., 2002). Two different reactions are catalyzed by glucansucrase enzymes, depending on the nature of the acceptor: i) hydrolysis, when water is used as acceptor; ii) glucosyl transfer (transferase), which can be divided in: a) polymerization, when the growing glucan chain is used as acceptor, and b) oligosaccharide synthesis, when oligosaccharides (e.g. maltose, isomaltose) are used as acceptor. Where studied, the linkage specificity of glucansucrases is conserved in oligosaccharide synthesis (Dols et al., 1997, Cote & Robyt, 1982, Robyt & Walseth,

Reuteransucrase of Lb. reuteri 121

53

1978), and oligosaccharides are elongated at their non-reducing end (Dols et al., 1997, Arguello Morales et al., 2001, Monchois et al., 2000a, Mukasa et al., 2000). A biochemical characterization of the reactions catalyzed by GTFA of Lb. reuteri 121 remained to be carried out. GTF proteins of lactic acid bacteria share a common structure and are composed of four distinct domains: their N-terminal end starts with (i) a signal peptide of 32-34 amino acids, followed by (ii) a highly variable stretch of 123-129 amino acids, (iii) a highly conserved catalytic or sucrose binding domain of about 1,000 amino acids and (iv) a C-terminal domain of about 500 amino acids, composed of a series of tandem repeats (Monchois et al., 1999d). We are interested in the precise roles of these various domains and repeats in the overall functioning of GTF enzymes. Amino acid sequence comparisons revealed that the catalytic core of GTF proteins is constituted of 8 β-sheets alternated with 8 α-helices (a (β/α)8 barrel structure) similar to glycoside hydrolases of family 13 (α-amylase family). This family includes for instance α-amylase and cyclodextrin glycosyltransferase (CGTase) (van der Veen et al., 2000). In GTFs, however, this (β/α)8 barrel structure is circularly permuted (MacGregor et al., 1996, Devulapalle et al., 1997). Therefore GTFs are classified in family 70 of glycoside hydrolases (http://afmb.cnrs-mrs.fr/CAZY/index.html). Whereas for the α-amylase family the catalytic mechanism is known (McCarter & Withers, 1994, Uitdehaag et al., 1999), the exact catalytic mechanism of GTF enzymes remains to be elucidated (Monchois et al., 1999d). Amino acid residues crucial for catalysis in glucansucrases of family 70 have been identified as Asp453 (putative catalytic nucleophile), Glu491 (putative acid/base catalyst) and Asp564 (putative transition state stabilizer) in GTFI from Streptococcus downei Mfe28 (Fig. 1) (MacGregor et al., 1996, Devulapalle et al., 1997). The equivalent invariable residues, in enzymes of α-amylase family 13, are Asp229, Glu257 and Asp328 (CGTase Bacillus circulans 251 numbering). In enzymes of both families the first mentioned Asp residue is involved in formation of the covalent glucosyl-enzyme complexes (Mooser et al., 1991, MacGregor et al., 1996, Uitdehaag et al., 1999). The importance of this Asp residue has also been shown for other GTFs by site-directed mutagenesis experiments (Kato et al., 1992, Monchois et al., 1997, Devulapalle et al., 1997) (Fig. 1). Based on alignments with glucansucrases from lactic acid bacteria (Monchois et al., 1999d), putative catalytic residues in Lb. reuteri strain 121 GTFA were identified and mutated (see results, Fig. 1). Different repeating units have been identified in the N-terminal variable domains of several glucansucrases: A-repeats in alternansucrase and dextransucrases from Leuconostoc mesenteroides sp. (Janecek et al., 2000), motif T in DSRT from Ln. mesenteroides NRRL B-512F (Funane et al., 2000), and motif S in DSRE from Ln. mesenteroides NRRL B-1299 (Bozonnet et al., 2002). The function of the N-terminal

Chapter 3

54

variable domain (and these repeats) has remained unclear. Deletion studies showed that it does not play a significant role in glucansucrase activity (Abo et al., 1991, Monchois et al., 1999a).

The C-terminal domains of Streptococcus and Leuconostoc GTF enzymes consist of a series of different tandem repeats, which have been divided into four classes: A, B, C and D repeats. DSRS from Ln. mesenteroides NRRL-512F contains in addition to A and C repeats also N repeats (Monchois et al., 1998b), which have not been identified in other GTFs. Alternansucrase from Ln. mesenteroides NRRL-B 1355 contains a single A repeat and distinct short repeats DG(X)4APY (Janecek et al., 2000).Within the A-D repeats, a repeating unit designated YG can be distinguished (Giffard & Jacques, 1994). The C-terminal repeats play an important role in glucan binding (Abo et al., 1991, Shah & Russell, 2002, Lis et al., 1995, Monchois et al., 1998b). The structure of GTFA of Lb. reuteri 121 is unusual because the above-mentioned repeats are not present in its N- and C-termini. Instead, GTFA possesses a relatively large N-terminal variable domain (702 amino acids) with 5 RDV repeats. Its relatively short C-terminal domain (267 amino acids) contains 11 YG-repeats only (Fig. 2A) (Kralj et al.,

Figure 1. Amino acid sequence alignment of highly conserved stretches (A, B, C) in catalytic domains of dextran-, mutan-, alternan and reuteransucrases of lactic acid bacteria (also see (Monchois et al., 1999d). GTFA, Lb. reuteri 121 (Kralj et al., 2002); GTFB, S. mutans GS5 (Shiroza et al., 1987); GTFD, S. mutans GS5 (Honda et al., 1990); GTFS, S. downei Mfe28 (Gilmore et al., 1990); GTFI, S. downei Mfe28 (Ferrettiet al., 1987); GTFJ, Streptococcus salivarius ATCC 25975 (Giffard et al., 1991); GTFK, S. salivarius ATCC 25975 (Giffard et al., 1993); DSRB, Ln. mesenteroides NRRL B-1299 (Monchois et al., 1998a); DSRS, Ln. mesenteroides NRRL B-512F (Monchois et al., 1997); ASR, Ln. mesenteroides NRRL B-1355 (Arguello-Morales et al., 2000); * , identical residue; : , highly conserved residue; . , conserved residue; ∀, putative nucleophile(MacGregor et al., 1996, Devulapalle et al., 1997); ⇓, putative acid/base catalyst (MacGregor et al., 1996, Devulapalle et al., 1997); ♦, putative residue stabilizing the transition state (MacGregor et al., 1996, Devulapalle et al., 1997); A, catalytic amino acids which have been mutated in other studies (Monchois et al., 1999d); A, catalytic amino acids of GTFA of Lb. reuteri 121 identified and characterized in this study.

Reuteransucrase of Lb. reuteri 121

55

2002). This raised questions about the precise role of the N- and C-terminal domains in GTFA.

Figure 2. Overview of primers and restriction sites used for cloning, expression and production of Lb. reuteri 121 GTFA protein and (deletion) mutant derivatives in E. coli and domain organization of GTFA and (deletion) mutant derivatives. (A) Overview of primers used for the Megaprimer method and the construction of p15GTFstop. The different domains shown in GTFA are i) N-terminal signal sequence (not present in mutants); ii) variable region with 5 RDV repeats (dark grey boxes); iii) catalytic domain; iv) C-terminal (putative) glucan binding domain with 4 YG-repeating units (dark grey boxes) according to the definition of (Giffard & Jacques, 1994) and 7 less conserved YG-repeating units (light grey boxes). (B) Schematic representation of the domain structure of GTFA-CHis and the truncated derivatives constructed. Molecular sizes, number of amino acids and position of the His tag of the different protein constructs are indicated.

Chapter 3

56

Here we describe the first molecular (construction of site-directed and deletions mutants) and biochemical (analysis of the main reactions catalyzed by wild type and mutant enzymes) characterization of an α-(1→4) synthesizing glucansucrase. Furthermore, we provide evidence for involvement of the C-terminal YG-repeats in glucan binding.

MATERIALS AND METHODS Bacterial strains, plasmids, media and growth conditions Escherichia coli DH5α (Phabagen) (Hanahan, 1983), E. coli TOP 10 (Invitrogen) and plasmid pBluescript II SK+ (Stratagene) were used for cloning purposes. Plasmid pET15b (Novagen) was used for expression of the (mutant) gtfA gene(s) in E. coli BL21 Star (DE3) (Invitrogen). E. coli strains were grown aerobically at 37 °C in LB medium (Ausubel et al., 1987). E. coli strains containing recombinant plasmids were cultivated in LB medium with the appropriate antibiotic (100 µg ml-1 ampicillin). Agar plates were made by adding 1.5 % agar to the LB medium. Molecular techniques General procedures for cloning, E. coli transformations, DNA manipulations, and agarose gel electrophoresis were as described (Sambrook et al., 1989). Restriction endonuclease digestions and ligations with T4 DNA ligase were performed as recommended by the enzyme suppliers (New England Biolabs; Roche Biochemicals). Primers were obtained from Eurogentec, Seraing, Belgium. Cycle sequencing (Murray, 1989) was performed on double stranded DNA using the Thermo Sequence fluorescent primer cycle sequence kit (Amersham Pharmacia Biotech). Sequence reactions were performed on the Amersham ALF-Express sequencing machine at the BioMedical Technology Center (Groningen, The Netherlands). DNA was amplified by PCR on a DNA Thermal Cycler PTC-200 (MJ Research) using Pwo DNA polymerase (Roche Biochemicals). Fragments were isolated from agarose gels using a Qiagen gel extraction kit (Qiagen) following the instructions of the supplier. Plasmid construction Plasmid pBGTF1, pBluescript carrying the Lb. reuteri 121 gene encoding GTFA (Kralj et al., 2002), without signal sequence and fused at the C-terminus with a stretch of six histidine residues (1,750 amino acids), was used. To facilitate future mutagenesis and nucleotide sequencing, the first of two EcoRV restriction sites (492 bp, 4105bp) was altered in pBGTF1, using the megaprimer method (Sarkar & Sommer, 1990) and the following primers: GTFpr1, 5-GATGCATGAGCTCCCATGGACCAACAAGTTCAG

Reuteransucrase of Lb. reuteri 121

57

CAAGCTTCC-3, containing SacI (italics) and NcoI (bold) restriction sites, and mEcoRV, 5-GGTTGTGATGTCACGCACAATGATTTG-3, containing a mutated EcoRV site (underlined, silent mutation by change of base shown in bold face). In a subsequent PCR reaction the amplified product (∼500 bp) was used as (forward) primer together with GTFpr12 5-GTGCATTAAAGTACGTAACCAATCAGTATTTCCGG-3, 1600 bp downstream of a XbaI restriction site (Fig. 2A). The resulting product of 2600 bp was digested with SacI and XbaI and ligated in the corresponding sites of pBGTF1, yielding pBGTF2. Plasmid pBGTF2 now consists (from 5 to 3) of four cassettes of 1000-1500 bp: i) SacI&(NcoI)/XbaI (1001 bp), ii) XbaI/PstI (1546 bp), iii) PstI/EcoRV (1454 bp), and iv) EcoRV/(BamHI)&ApaI (1269 bp) (Fig 2A). This plasmid was digested with PstI (2655 bp) and EcoRV (4105 bp) and the resulting fragment (1454 bp) was ligated in the corresponding sites of pBluescript II SK+ (Stratagene), yielding pBPE1500. This (small) construct was used for site-directed mutagenesis, sequencing and rapid exchange (using PstI and EcoRV restriction sites) with pBGTF2 (see below). pBGTF2 was also digested with NcoI/BamHI (5257 bp) and ligated in the corresponding sites of pET15b (Novagen), yielding p15GTF2 (encoding GTFA-CHis, Fig. 2B). This construct was used in subsequent cloning steps. In order to create a construct without C-terminal His-tag, but with a stop codon, the following primers were used to exchange the C-terminal part of p15GTF2: SpeFor: 5-GGGCTTTCAGATGCAACTAATCGTTGGGG-3, lying 400 bp upstream of a SpeI restriction site, and GtfARevstop 5-TCGATGGGCCCCGGATCCTATTATAGTTTAT TTTGATCAAGCATCTTACC-3, containing BamHI (bold) and ApaI (italics) restriction sites (Fig. 2A). The resulting PCR product (∼2000 bp) was digested with SpeI and BamHI (1592 bp) and ligated in the corresponding sites of p15GTF2 (see above), yielding p15GTFstop. In all cases successful mutagenesis was confirmed by nucleotide sequencing. Site-directed mutagenesis of putative catalytic residues of GTFA Plasmid pBPE1500, (see above), was used as template for mutagenesis. The QuickChangeTM site-directed mutagenesis kit (Stratagene) was used to construct mutants D1024N, E1061Q and D1133N, using different primer pairs (Table 1A). Successful mutagenesis resulted in the appearance or disappearance of restriction sites, allowing rapid screening of potential mutants. After successful mutagenesis (confirmed by nucleotide sequencing), pBPE1500 (containing mutation) was digested with PstI and EcoRV and ligated in the corresponding sites of pBGTF2. The resulting plasmid pBGTF2 (containing mutation) was digested with NcoI/BamHI and the resulting 5.3 kb fragment was ligated into the corresponding sites of the expression vector pET15b (Novagen) yielding p15GTF2 containing mutation.

Chapter 3

58

Construction of truncated gtfA genes Appropriate primer pairs and template DNA was used (Table 1B) to construct eight different deletion mutants of gtfA (Fig. 2B). Deletion mutants were constructed by exchange of restriction fragments and checked by nucleotide sequencing.

Purification of GTFA (mutant) proteins Cells of E. coli BL21star (DE3) harboring the different pET15b derivatives were harvested by centrifugation (10 min at 4 °C at 10,000 × g) after 16 h of growth at 37 °C (without induction). The pellet was washed with 50 mM phosphate buffer pH 8.0. Pelleted cells were resuspended in 50 mM sodium phosphate buffer pH 8.0 containing 250 mM NaCl, 5 mM β-mercaptoethanol and 10 mM imidazole. Cells were broken by sonication (7 × 15 sec at 7 micron with 30 sec intervals) and centrifuged (10 min at 4 °C at 10,000 × g). The clear lysate containing GTF activity was loaded on a Ni-NTA column

Table 1. Oligonucleotides used for site-directed (A) and deletion mutagenesis (B) of gtfA. Nucleotides in bold type/italics represent mismatches with the sequence of gtfA. Nucleotides underlined represent introduced or removed (in the case of mutation E1061Q) restriction sites. H = histidine residue, st = stop codon.

Reuteransucrase of Lb. reuteri 121

59

(Qiagen). Binding was realized using 50 mM sodium phosphate buffer pH 8.0 containing 250 mM NaCl, 5 mM β-mercaptoethanol and 10 mM imidazole, followed by washing using the same buffer. Elution of His-tagged protein(s) was performed using 50 mM sodium phosphate buffer pH 8.0 with 250 mM NaCl, 1 mM β-mercaptoethanol and 200 mM imidazole. Eluted proteins were desalted with 20 mM Tris buffer, pH 8.0, using a 5 ml Hi-Trap desalting column (Amersham Pharmacia Biotech). Subsequently, the samples were purified on an ACTA prime FPLC system (Amersham Pharmacia Biotech), using a 1 ml Resource-Q column (Amersham Pharmacia Biotech) and a linear gradient of 30 ml with 1 M NaCl in 20 mM Tris buffer pH 8.0 as eluens at a flow rate of 1 ml.min-1. Proteins present in the elution peak were desalted with 25 mM sodium acetate buffer pH 4.7, supplemented with 1 mM CaCl2, using a 5 ml Hi-Trap desalting column (Amersham Pharmacia Biotech). At each stage of the purification, the GTF transferase activity was quantified as previously described (van Geel-Schutten et al., 1999). The degree of purity of the different mutants was determined by SDS-PAGE (Laemmli, 1970). Protein concentrations were determined using the Bradford method using the Bio-Rad reagent and BSA (bovine serum albumin) as a standard (Bio-Rad). From 250 ml cultures of E. coli strains expressing full-length (wild type or mutant) and N-or C-terminally truncated protein, approximately 0.25 and 1 mg of highly purified protein was obtained, respectively. Glucan binding assay The glucan binding ability of GTFA (deletion mutants) was measured as described previously (Lis et al., 1995), and as improved by (Shah & Russell, 2002). Briefly, 30 pmol of GTFA (mutant) protein in 200 µl PBS-T (20 mM sodium phosphate pH 7.3, 150 mM NaCl, 0.05% tween 20) was incubated overnight at 4°C in Ni-NTA coated HisSorb plates (Qiagen). Proteins containing a His-tag are bound to the Ni-NTA, non-binding protein was removed by washing with PBS-T (four times 1 min). After incubation for 20 min with 200 µl biotin labelled dextran (100 µg/µl; Fluka) in PBS-B (20 mM sodium phosphate pH 7.3, 150 mM NaCl, 0.2% BSA), which may bind to the glucan-binding domain (GBD) of the protein, washing was performed with PBS-T, as described before. Extravidin-alkaline phosphatase conjugate (1/10,000; Sigma) in PBS-B, which binds to the biotin part of the labelled dextran, was added and incubated for 1 h, followed by washing with PBS-T. Subsequently, alkaline phosphatase substrate 4-nitrophenyl phosphate (100 µl; Roche Biochemicals) was added, yielding a yellow color upon its hydrolysis by the bound alkaline phosphatase. The colour change was monitored using a SpectraMax Plus 384 plate reader (Molecular Devices).

Chapter 3

60

Enzyme activity assays The various reuteransucrase activities were determined by measuring glucose and fructose release (enzymatically) from sucrose conversion (van Geel-Schutten et al., 1999). The amount of fructose released (VF) corresponds to total enzyme activity (initial formation of leucrose or other sucrose isomers was negligible, see results and data not shown). The amount of free glucose (VG) represents the hydrolytic activity of the enzyme. The amount of fructose minus the amount of free glucose reflects the transferase activity (VF-VG). Unless indicated otherwise, reactions were performed at 50 °C in 25 mM NaAc buffer, pH 4.7, containing 1 mM CaCl2 and 30 nM purified (mutant) reuteransucrase enzyme. One unit of enzyme activity is defined as the release of 1 µmol of monosaccharide per min. In case of very low activity of mutant proteins, assay conditions were modified as follows: reactions were performed with 25 × more protein, and aliquots were removed at 30 min intervals instead of 1 min. (i) Kinetic parameters. Kinetic assays were performed using twenty-four different sucrose concentrations ranging from 0.25 to 100 mM. Over a 6 min incubation period, samples of 25 µl were withdrawn every minute and inactivated with 2.5 µl 1 M NaOH. Curve fitting of the data was performed with the SigmaPlot program (version 8.0) using either, the Michaelis-Menten formula [ y = (a × x) / (b + x) ], or the same formula with a substrate inhibition constant [ y = (a × x) / (b + x + (x 2/c))]. In these equations y is the specific activity (U.mg-1), x is the substrate concentration (mM sucrose), a is the maximal reaction rate, Vmax (U.mg-1), b is the affinity constant for the substrate (Km, mM sucrose), and c is the substrate inhibition constant (Ki, mM sucrose). (ii) Effect of maltose on initial GTFA activity. The initial rate of oligosaccharide synthesis was examined by measuring the effect of maltose on (mutant) GTFA enzyme activity, using 50 mM sucrose and 100 mM maltose. The activity was determined by measuring fructose release. Product analysis (i) Product spectrum from sucrose. After complete depletion of sucrose (100 mM, 60 h at 50 °C), the concentrations of fructose, glucose and leucrose in the reaction medium of GTFA (mutants) were determined using anion exchange chromatography (see below). The amount of fructose released (97.9 %) and leucrose (2.1 %) synthesized from sucrose corresponds to 100 %. Subtracting the free glucose (23.1 %; due to hydrolysis) from the free fructose (97.9 %) concentration, allowed calculation of the yield of reuteran synthesis (74.8 %) from sucrose (data of GTFA-CHis used here as an example, see Table 4).

Reuteransucrase of Lb. reuteri 121

61

(ii) Oligosaccharide synthesis with maltose and isomaltose as acceptors. Oligosaccharide synthesis was analyzed using 100 mM sucrose together with maltose or isomaltose (100 mM each). After complete consumption of sucrose (60 h at 50 °C), samples were diluted 500-1000 times in a 90% DMSO solution. Maltose, isomaltose, maltotriose, panose (Sigma), isomaltotriose (TNO Nutrition and Food Research, Groningen, The Netherlands), sucrose (Acros Organics), fructose, glucose (Merck), and leucrose (Pfeiffer & Langen) were used as standards. The percentage of oligosaccharide synthesis from sucrose and acceptor was determined by subtracting the amount of unused acceptor from the initial acceptor concentration. Separation of oligosaccharides was achieved with a CarboPac PA1 anion exchange column (250 × 4 mm; Dionex) coupled to a CarboPac1 guard column (Dionex). The following gradient was used: eluent A (0 min, 100%); (5 min, 100%); (50 min, 92%); (55-58 min, 0%); (60 min, 100%); (75 min, 100%). Eluent A was sodium hydroxide (0.1 M) and eluent B was NaAc (0.6 M) in sodium hydroxide (0.1 M). Detection was performed with an ED40 Electrochemical detector (Dionex) with an Au working electrode and an Ag/AgCl reference electrode with a sensitivity of 300 nC. The pulse program used was: +1.0 Volt (0-0.40s); +0.7 Volt (0.41-0.60 s); -0.1 Volt (0.61-1.00 s). Data were integrated using a Turbochrom (Applied Biosystems) data integration system. To determine the degree of polymerization (DP) of unknown oligosaccharides a BC-200 Ca2+ column (at 85 °C; 300 by 7.8 mm; Benson Polymeric) eluted with water (0.2 ml/min) was used. Detection was done by using a model 830-RI refractive index detector at 40 °C (Jasco).

In vitro glucan production by (mutant) GTFA enzymes and glucan structure analysis (i) Polymer production. Purified (mutant) enzyme preparations were incubated overnight with 146 mM sucrose, using the conditions described above under enzyme assays. Glucans produced were isolated by precipitation with ethanol (van Geel-Schutten et al., 1999). (ii) Methylation analysis. Polysaccharides were permethylated using methyl iodide and Na-Dimsyl in DMSO at room temperature (Hakomori, 1964). After hydrolysis with 2 M trifluoroacetic acid (1 h, 125 °C), the partially methylated monosaccharides were reduced with NaBD4 (Harris et al., 1984). Mixtures of partially methylated alditol acetates obtained were analyzed by GLC on a CP Sil 5 CB column (25 m × 0.53 mm; Chrompack) and by GLC-mass spectrometry (MS) on a RTX 5 Sil MS (30 m × 0.25 mm; Restek) column (Chaplin, 1982, Jansson et al., 1976).

Chapter 3

62

(iii) Molecular masses of the glucans. Molecular mass analysis was performed as described previously, using high performance size exclusion chromatography (HPSEC) coupled on-line with a multi angle laser light scattering (MALLS) and differential refractive index detection (Kralj et al., 2002). RESULTS Purification of (mutant) GTFA enzymes Lb. reuteri 121 GTFA-CHis enzyme expressed in E. coli was purified to homogenity. Also three site-directed (D1024N, E1061Q and D1133N; Fig. 1), and eight deletion mutants (GTFA-∆N-CHis, GTFA-∆N-NHis, GTFA-∆N∆4YG-NHis, GTFA-∆N∆5YG-NHis, GTFA-∆N∆6YG-NHis, GTFA-∆N∆7YG-NHis, GTFA-∆N∆8YG-NHis, and GTFA-∆N∆11YG-NHis; Fig. 2B) were expressed in E. coli and purified to homogenity. The predicted Mr of the different deletion variants (Fig. 2B) was in agreement with the results obtained by SDS-PAGE analysis (data not shown). Effects of pH, temperature and metal ions on GTFA activity In order to define the best conditions for subsequent kinetic studies, the pH and temperature optima of GTFA activity were examined. The pH optimum for the

Figure 3. The effect of pH on GTFA activity (A). Enzyme activity was determined at 50 °C in the presence of 1 mM CaCl2 by measuring the amount of glucose and fructose released in 30 min from 50 mM sucrose by 30 nM GTFA-CHis (means ± S.E.M.; n = 3). Solid line: Transferase activity. Dashed line: Hydrolysis activity. (!) 25 mM Potassium acetate buffer; (!) 25 mM MES buffer; (") 25 mM MOPS buffer. The effect of temperature (B) on transferase activity of GTFA was evaluated as described for Figure 3A (means ± S.E.M.; n = 3).

Reuteransucrase of Lb. reuteri 121

63

hydrolyzing activity was in the range of pH 4.5 to 5.5 (Fig. 3A). For the transferase activity a pH optimum at pH 4.7 was observed. The temperature optimum for both reactions was 50 °C. At lower temperatures activity remained relatively high. (Fig. 3B, only shown for transferase activity). Different metal ions had strongly varying effects on the hydrolytic and transferase activities of GTFA (Table 2). The Zn2+, Hg2+, Fe3+, Fe2+ and Cu2+ ions significantly inhibited both enzyme activities, but especially the transferase activity, which was not detectable in their presence. EDTA also inhibited transferase activity completely, but hydrolysis remained almost unaffected. In contrast, Ca2+, K+, Na2+ and Mg2+ ions had stimulating effects on GTFA enzyme activities. Of these compounds Ca2+ ions had the most stimulating effect, with hydrolysis and transferase activities increased 2 and 8-fold, respectively.

Kinetic studies of GTFA (i) Kinetic parameters. In the presence of sucrose GTFA-CHis displayed Michaelis-Menten type of kinetics for the transferase reaction (VF-VG) and for total enzyme activity (VF). The hydrolysis reaction (VG) displayed Michaelis-Menten with substrate inhibition (Fig. 4A; Table 3). The initial hydrolysis rate represents 95 % of the initial sucrose consumption rate in the presence of 0.5 mM sucrose, 75 % with 2.5 mM sucrose and only 65 % with 10 mM sucrose. These data reveal that GTFA-CHis favours hydrolysis at low sucrose concentrations and polymerization at high sucrose concentrations. (ii) Effect of maltose on initial GTFA activity. Maltose stimulated the GTFA-CHis total activity (VF) and transferase activity (VF-VG) 3.5 and 7.5 fold, respectively. The hydrolysis (VG) rate decreased 3.5 fold when maltose was present (Table 3B, data not shown).

Table 2. Effects of various compounds on GTFA-CHis catalyzed activities. Activity measurements were done at 50 °C in a 25 mM NaAc buffer (pH 4.7), with 50 mM sucrose. Glucose release (VG, hydrolysis) and fructose minus glucose release (VF-VG, transferase) from sucrose by 30 nM GTFA-CHis after 30 min of incubation was taken as a measure for enzymatic activity. Results are given as means ± S.E.M.; n = 3. ND, Transferase activity could not be detected

Chapter 3

64

Product analysis (i) Product spectrum from sucrose. After complete depletion of sucrose, GTFA-CHis showed the following product distribution: reuteran synthesis, 74.8 %; hydrolysis, 23.1 %; and leucrose (α-D-glucopyranosyl-(1→5)-β-D-fructofuranoside) synthesis, 2.1 % (Table 4). Dionex analysis showed that a minor amount of isomaltose and significant amounts of an unknown product that eluted after 35 min were synthesized (data not shown). (ii) Oligosaccharide synthesis with maltose and isomaltose as acceptor substrates. Dionex analysis showed that, in the presence of maltose, GTFA-CHis formed panose (α-D-glucopyranosyl-(1→6)-α-D-glucopyranosyl-(1→4)-D-glucose) as most abundant acceptor reaction product (from 100 mM sucrose and 100 mM maltose, approximately 44 mM panose was synthesized), indicating that an α-(1→6) linkage was formed at the non-reducing end of maltose (Fig. 5B). When isomaltose was used as acceptor low amounts of isomaltotriose and isomaltotetraose (not shown in standard; elution between 35-36 min) were formed, together with two abundant unknown oligosaccharide products (Fig. 5C) with a degree of polymerization (DP) of 3 (most likely isopanose; α-D-glucopyranosyl-(1→4)-[α-D-glucopyranosyl-(1→6]-D-glucose) and 4, respectively (DP determined by HPLC, data not shown).

Figure 4. Effect of sucrose concentration on initial GTFA-CHis enzyme activity at 50 °C in 25 mM NaAc buffer pH 4.7 supplemented with 1 mM CaCl2 (A). Effect of sucrose concentration on initial GTFA-∆N∆6YG-NHis enzyme activity (B). The mutants GTFA-∆N∆4YG-NHis and GTFA-∆N∆5YG-NHis also displayed the latter type of kinetics. The reactions were performed with 30 nM enzyme. (!) VG (hydrolytic activity), (#) VG-VF (transferase activity), (!) VF (total activity).

Reuteransucrase of Lb. reuteri 121

65

Characteristics of mutants in the putative catalytic amino acid residues Based on alignments of GTFA with different GTF enzymes, three (putative) catalytic residues were identified in GTFA and mutated (Fig. 1). In the E. coli host strain, GTFA mutants D1024N, E1061Q and D1133N were expressed to similar levels as the wild type enzyme (data not shown). Analysis of the purified proteins showed that they had almost completely lost their activity. Compared to wild type, D1024N and E1061Q showed a 1,000 fold reduction of total enzyme activity (VF). Mutation D1133N resulted in an enzyme with a 300 fold reduced total activity (VF).

Characteristics of GTFA deletion mutants (i) Glucan binding. GTFA-∆N-CHis displayed lower glucan binding activity (GBA) than GTFA-∆N-NHis (Figs. 2B and 6). GTFA-CHis showed virtually no GBA. GTFA-∆N∆4YG-NHis completely lacked GBA. GTFA-∆N∆5YG-NHis, GTFA-∆N∆6YG-NHis and GTFA-∆N∆7YG-NHis were active again but showed gradually lower GBA (Fig. 6). GTFA-∆N∆8YG-NHis and GTFA-∆N∆11YG-NHis were virtually inactive. (ii) Kinetic analysis. GTFA-∆N-NHis and GTFA-∆N-CHis (Fig. 2B) displayed Michaelis-Menten type of kinetics in all three reactions. Interestingly, both mutants had a 3-4 fold increased transferase activity, whereas their hydrolytic activities decreased (Table 3). The Michaelis-Menten formula with a substrate inhibition constant was used to fit the

Table 3. Kinetic parameters of wild type and deletion mutants of Lb. reuteri 121 GTFA: (A) kinetic parameters were determined using 24 different concentrations of sucrose (0.25-100 mM), (B) initial total activity determined using 50 mM sucrose and 100 mM maltose. Kinetic parameters for GTFA-∆N-∆7YG-NHis, GTFA-∆N-∆8YG-NHis and GTFA-∆N-∆11YG-NHis were not determined due to their low activity (total activity 43 ×, 63 × and 1000 × reduced, respectively, compared to wild type).*, Affinity for sucrose could not be determined when measuring transferase and total activity of GTFA-∆N-∆4YG-NHis, GTFA-∆N-∆5YG-NHis and GTFA-∆N-∆6YG-NHis, due to failure to reach saturation of the enzyme at the sucrose concentrations used, resulting in high standard errors with curve fits (see also Fig. 4B). Therefore activities at a substrate concentration of 50 mM sucrose are depicted. Data obtained were fitted using the Michaelis-Menten equation (with or without (#) substrate inhibition).

Chapter 3

66

hydrolytic activity of the mutants GTFA-∆N-∆4YG-NHis, GTFA-∆N-∆5YG-NHis and GTFA-∆N-∆6YG-NHis. The transferase and total activity of these mutants could not be fitted with the Michaelis-Menten formula. The affinity for sucrose had decreased drastically in both reactions and no saturation was reached with the sucrose concentrations used (Fig. 4B). Sequential deletions of the YG-repeats from the C-terminus onwards resulted in initially lower hydrolytic activity compared to the wild type enzyme (Table 3). However, the affinity for sucrose in the hydrolysis reaction was comparable to wild type (Table 3). Kinetic parameters for mutants GTFA-∆N-∆7YG-NHis, GTFA-∆N-∆8YG-NHis and GTFA-∆N-∆11YG-NHis were not determined due to their low total activity (43 ×, 63 × and 1000 × reduced, respectively, compared to wild type levels).

(iii) Effect of maltose on initial mutant GTFA activity. The maltose stimulatory effect on GTFA-∆N-NHis and GTFA-∆N-CHis total activity was 1.5 to 2 times lower than on wild type. Total activity of mutants GTFA-∆N-∆4YG-NHis, GTFA-∆N-∆5YG-NHis and GTFA-∆N-∆6YG-NHis on the contrary was stimulated 5.5, 6 and, 7.5 times, respectively, by maltose (Table 3B). (iv) Product spectrum from sucrose. When incubated with sucrose, mutants with C-terminal deletions showed slightly increased reuteran synthesis and (slightly) decreased hydrolysis (Table 4).

Table 4. Effects of deletions in Lb. reuteri 121 GTFA on the product spectrum obtained with sucrose (100 mM), or sucrose and maltose (100 mM each), or sucrose and isomaltose (100 mM each). *Sucrose was not completely consumed. Oligosaccharide yield is expressed as percentage of total amount of acceptor used in incubation.

Reuteransucrase of Lb. reuteri 121

67

Figure 5. Dionex analysis of Lb. reuteri 121 GTFA acceptor reaction products (as described in materials and methods). A) Elution profile of a standard mixture, B) Products formed upon incubation of 30 nM GTFA-CHis enzyme with 100 mM sucrose and 100 mM maltose for 60 h, C) Products formed upon incubation of 30 nM GTFA-CHis enzyme with 100 mM sucrose and 100 mM isomaltose for 60 h.

Chapter 3

68

(v) Oligosaccharide synthesis with maltose and isomaltose as acceptor substrates. In the acceptor reaction, all deletion mutants showed similar oligosaccharide yields as wild type (Table 4). The deletion mutants produced the same oligosaccharides as wild type (Fig. 5, data not shown).

(vi) Molecular sizes and linkage type analysis of glucans produced. Compared to GTFA-CHis, the deletion mutants showed no drastic changes in the sizes and linkage types of the glucans produced (Table 5). However, the glucan products of mutants GTFA-∆N∆5YG-NHis and GTFA-∆N∆6YG-NHis showed less branching (lower percentage of

Figure 6. Glucan binding activity of the different Lb. reuteri 121 GTFA deletion mutants, measured after 5.5 h of addition of substrate(means ± S.E.M.; n = 3). Protein numbers refer to the various (deletion) mutants shown in Fig. 2B.

Table 5. Methylation analysis and molecular masses of the glucans produced by purified wild type Lb. reuteri 121 GTFA enzyme and derived deletion mutants.

Reuteransucrase of Lb. reuteri 121

69

both terminal and branched, α-(1→4,6) glucosyl units), a higher percentage of α-(1→6) linkages, and slightly decreased molecular masses compared to wild type enzyme and other deletion mutants (Table 5). DISCUSSION Characterization of GTFA This paper reports a detailed biochemical and molecular characterization of the first recombinant glucansucrase (GTFA) of a Lactobacillus strain, synthesizing a polymer with mainly α-(1→4) glucosidic linkages (a reuteransucrase). The GTFA pH optimum was comparable to that of other glucansucrases. Interestingly, transferase activity showed a sharp optimum around pH 4.7 whereas the hydrolysis reaction had a broader optimal pH range (Fig. 3A). This is the first time demonstration of different pH profiles for these GTF catalyzed reactions. A striking feature of the recombinant GTFA-CHis protein is its high optimal temperature of 50 °C (Fig. 3B). This high temperature optimum was also observed for the inulosucrase and levansucrase enzymes of the same strain (van Hijum et al., 2002, van Hijum et al., 2003, van Hijum et al., 2004). Of all cat-ions tested Ca2+ had the most stimulating effect on enzyme activity (Table 2): hydrolysis and transferase activity were increased 2 and 8-fold, respectively. The Zn2+, Hg2+, Fe3+, Fe2+ and Cu2+ ions significantly inhibited enzyme activity. These compounds (except for Hg2+ which was not tested) also inhibited the gtfB, gtfC and gtfD encoded enzymes of Streptococcus mutans (Wunder & Bowen, 1999). Kinetics of GTFA Kinetic analysis of GTFA-CHis revealed interesting features (Fig. 4). At high sucrose concentrations, substrate inhibition was observed for the GTFA-CHis hydrolysis reaction. GTFA-CHis clearly favours hydrolysis at low sucrose concentrations and transferase at high sucrose concentrations. The same phenomenon has otherwise been observed only for the family 13 enzyme amylosucrase from Neisseria polysaccharea (Potocki de Montalk et al., 2000). The GTFA affinity for sucrose in the total reaction was much higher (Km of 0.9 mM) than observed for GTFI of S. downei Mfe28 and DSRS of Ln. mesenteroides NRRL B-512F (Km values of 49 and 26 mM, respectively) (Monchois et al., 2000a, Monchois et al., 1997). The stimulating effect of maltose on initial GTFA velocity (3-4 fold activation of total activity) was also observed for DSRS of Ln. mesenteroides NRRL B-512F and GTFI of S. downei Mfe28 (2- and 4 fold stimulation of total activity, respectively) (Monchois et al., 2000a, Monchois et al., 1997).

Chapter 3

70

Product spectrum from sucrose Upon depletion of sucrose, GTFA had formed only low levels of leucrose (2.1 % of total sucrose consumed), even when incubated together with a high concentration of fructose 100 mM (about 5 % leucrose synthesized) (Table 4; data not shown). GTFI of S. downei on the contrary produced significant amounts of leucrose (24.0 %) (Monchois et al., 2000b, Monchois et al., 2000a). GTFA synthesized more glucan (74.8 %), and was more effective in hydrolysis (23.1 %) than GTFI of S. downei Mfe28 (60.0 % and 16.2 %, respectively) (Monchois et al., 2000a). Fructose thus is a better acceptor for GTFI of S. downei Mfe28 than for GTFA of Lb. reuteri 121. Oligosaccharide synthesis with maltose and isomaltose as acceptors Analysis of the acceptor reaction of GTFA revealed that with glucose and sucrose only a small amount of isomaltose was formed (data not shown). Previous studies showed that the production of isomaltose by DSRS from Ln. mesenteroides NRRL B-512F could be improved by using high amounts of sucrose and glucose (Buchholz & Seibel, 2003). Maltose is a good acceptor reaction substrate for DSRS of Ln. mesenteroides B-512F (Robyt & Walseth, 1978, Monchois et al., 1997), for GTFI of S. downei Mfe28 (Monchois et al., 1999c), and for GTFA of Lb. reuteri 121 (this study). GTFA produced panose as major product with maltose as acceptor (formation of an α-(1→6) linkage to the non-reducing end of maltose), together with low amounts of maltotriose and two unknown products (Fig. 5B). The formation of panose has also been observed for other GTF enzymes (Monchois et al., 1997, Monchois et al., 1998a, Monchois et al., 1996, Monchois et al., 2000b, Robyt, 1996). Compared to GTFI of S. downei Mfe28 (91.3 %) and DSRS of Ln. mesenteroides NRRL-B512F (93 %) (Monchois et al., 2000a, Monchois et al., 1998b), the yield of oligosaccharides synthesized by GTFA-CHis with maltose as acceptor substrate (59.6 %) is lower. Maltose thus is a better acceptor reaction substrate for GTFI and DSRS than for GTFA. When isomaltose was used as acceptor reaction substrate, two unknown products (identified by HPLC as DP 3 and DP 4, data not shown) were produced. Maltotriose or panose eluted at different time points than the DP 3 oligosaccharide; its most likely identity is isopanose. The coupling of a glucose moiety, using one of the two linkage types synthesized by GTFA, α-(1→4) and α-(1→6), to the reducing end of isomaltose would have resulted in the formation of isomaltotriose (of which small amounts are indeed synthesized), or panose (not formed) (Fig. 5C). However, various studies show that glucansucrases elongate oligosaccharides at their non-reducing end (Mukasa et al., 2000, Dols et al., 1997, Arguello Morales et al., 2001, Monchois et al., 2000a). The small amount of isomaltotriose is thus probably formed by coupling of glucose with an α-(1→6) linkage to the non-reducing end of isomaltose. Linkage specificity (inside glucans) of several glucansucrases is conserved in oligosaccharide synthesis (Dols et al., 1997, Cote

Reuteransucrase of Lb. reuteri 121

71

& Robyt, 1982, Robyt & Walseth, 1978). This suggests that the DP 3 oligosaccharide is isopanose (coupling of a glucose unit by an α-(1→4) glucosidic bond onto the non-reducing end of isomaltose). Besides the DP 3 oligosaccharide, also a DP 4 oligosaccharide was synthesized. The exact nature of both oligosaccharides will be analyzed in future. However, it appears likely that reuteransucrase is capable of forming both α-(1→4) and α-(1→6) glucosidic bonds with acceptor reaction substrates (Fig. 5). Characteristics of catalytic mutants Mutagenesis of the Lb. reuteri strain 121 GTFA putative catalytic residues (mutants D1024N, E1061Q and D1133N) resulted in drastically reduced total enzymatic activities. D1024 of GTFA is homologous to: 1) the Asp residues identified by Mooser et al. (Mooser et al., 1991) as part of the active site of Streptococcus sobrinus GTFI and GTFS, respectively, 2) Asp451 of GTFB from S. mutans GS5, 3) Asp453 of GTFI from S. downei Mfe28 and 4) Asp551 of DSRS of Ln. mesenteroides NRRL-B512F, which were also shown to be essential for activity (Kato et al., 1992, Monchois et al., 1997, Devulapalle et al., 1997) (Fig. 1). GTFA residues E1061 and D1133 correspond to residues E491 and D564 in GTFI of S. downei Mfe28. These residues are essential for activity in both GTFI and GTFA (Devulapalle et al., 1997); this study) Characteristics of deletion mutants GTFA possesses a relatively large N-terminal variable domain (702 amino acids) containing 5 RDV repeats (Kralj et al., 2002). Its complete deletion had a drastic effect on enzyme kinetics. The initial transferase activity of GTFA-∆N-CHis and GTFA-∆N-NHis increased 3-4 fold, whereas the hydrolytic activity decreased (Table 3). Conceivably, the large N-terminal domain in the wild type protein causes steric hindrance to the growing glucan chain, its deletion resulting in a strongly increased transferase activity. The affinities for the substrate sucrose remained rather similar, in both the transferase and the hydrolysis reaction (Table 3). The data show that the large N-terminal variable domain is important for initial activity with sucrose (Table 3), but has only small effects on the product spectrum with sucrose alone, or sucrose plus acceptor substrates (Table 4), and on glucan characteristics (Table 5). The GTFA mutants with N-terminal and additional C-terminal deletions showed drastically decreased affinity for sucrose in the total (VF) and transferase (VF-VG) reactions (Fig. 4B; Table 3). In contrast, C-terminal deletions in DSRS of Ln. mesenteroides NRRL B-512F had virtually no effects on Km values with sucrose (Monchois et al., 1998b). Sequential deletions of the YG-repeats from the C-terminus onwards resulted in gradually lower activity of the mutant GTFA enzymes in all reactions (Table 3A). The stimulatory effect of maltose increased upon deletion of C-terminal YG-repeats (Table 3B). Both phenomena were also observed for DSRS from Ln. mesenteroides NRRL B-512F

Chapter 3

72

(Monchois et al., 1998b). The increased stimulatory effect by maltose upon C-terminal deletions in DSRS was explained by a change in a rate-limiting reaction step. C-terminal deletions in GTFI from S. downei Mfe28 had no effect on distribution of glucosyl residues during mutan synthesis (Monchois et al., 1999a). The reuterans produced by wild type GTFA and C-terminal deletion mutants derived had comparable sizes and linkage type distribution, except for GTFA-∆N∆5YG-NHis and GTFA-∆N∆6YG-NHis. The latter mutants showed less branching (lower amount of both terminal and branched, α-(1→4,6) glucosyl units), a higher percentage of α-(1→6) linkages and a slightly decreased molecular mass compared to the wild type and other mutants (Table 5). C-terminal deletions in GTFI of S. downei Mfe28 did not affect glucan structure (Monchois et al., 1999a). However, deletion of six internal units (A-C) of GTFG of Streptococcus gordonii also affected glucan structure (Vickerman et al., 1996). GTFA enzyme activity could still be detected after removal of more than half (152 amino acids, 6-YG-repeats) of the GBD (Table 3). When more than 6-YG-repeats were removed enzyme activity was reduced drastically (Table 3). C-terminal deletions in other glucansucrases also affected enzyme activity (Monchois et al., 1998b, Kato & Kuramitsu, 1990). Only in case of GTFI of S. downei Mfe28 the C-terminal domain could be removed completely without large loss of activity (Monchois et al., 1999a). The C-terminal domains of glucansucrases from leuconostocs and streptococci are involved in glucan binding (Abo et al., 1991, Shah & Russell, 2002, Lis et al., 1995, Monchois et al., 1998b). The A repeats found in the C-terminal domains of these GTF enzymes are not present Lb. reuteri 121 GTFA. Instead, 11 YG-repeats were identified (Kralj et al., 2002). Their deletion revealed that the GTFA C-terminal domain and these YG-repeats are involved in glucan binding (Fig. 6). GTFA-∆N∆4YG-NHis was not able to bind biotin-dextran, which may be due to conformational changes in this mutant protein. GTFA-CHis showed virtually no GBA. The location of the His-tag (GTFA-∆N-CHis also showed lower GBA than GTFA-∆N-NHis), together with the presence of the large N-terminal variable domain, may block the access of the GBD to the biotin-dextran. We had to incubate the GTFA protein / biotin-dextran / extravidin-alkaline phosphatase complex for several hours with the 4-nitrophenyl phosphate substrate to be able to measure GBA, whereas in a previous report, with GFTI of S. downei Mfe28, incubation for a few minutes was enough to measure comparable amounts of GBA (Shah & Russell, 2002). Reuteran rather than biotin-dextran may be the preferred ligand for the C-terminal repeats of GTFA. Nevertheless, we show for the first time that also in reuteransucrase, an α-(1→4) synthesizing glucansucrase from a Lactobacillus strain, the C-terminal domain is involved in glucan binding. Furthermore, positions at which truncation prevented glucan binding (deletion of 8 YG-repeats; 100 amino acids after the end of catalytic core) were identified (Fig. 6).

Reuteransucrase of Lb. reuteri 121

73

The GTFA hydrolytic activity remained upon deletion of YG-repeats, whereas glucan binding and affinity for sucrose in the transferase reaction decreased. This may imply that the glucan binding domain is involved in polymer chain growth, as previously postulated for the GTFI from S. mutans GS5 and GTFI from S. sobrinus 6715 (Kato & Kuramitsu, 1990, Abo et al., 1991). Conclusion A thorough molecular and biochemical investigation of GTFA, the first GTF enzyme synthesizing linkages of the α-(1→4) glucosidic type, has been carried out. This includes analysis of the kinetics of wild type enzyme and N- and C-terminally truncated mutants derived, of the acceptor reaction, and of glucan binding by the C-terminal domain. Finally three catalytically important residues have been identified and characterized. ACKNOWLEDGEMENTS We thank Peter Sanders for help with the Dionex and GC analysis and the kind gift of isomaltotriose, Wim Soetaert for the kind gift of leucrose, Albert Woortman and Peter Steeneken for help with the methylation analysis, Gert-Jan Euverink for help with enzyme purification and assays, Elly Faber for GC-MS analysis and Kees de Ruijter for the HPSEC-MALLS analysis.